INDEX

Explanations

punctuation marks

oai_token-act-pair · gpt-4o-mini Triggered by @bot

introduces an assertion or observation

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.31.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

parsedMessage

-0.65

出版年

-0.64

 GenerationType

-0.59

 useAppContext

-0.57

RegistryLite

-0.57

DockStyle

-0.56

 Normdatei

-0.56

ok

-0.56

-0.55

BeginContext

-0.54

POSITIVE LOGITS

it

0.66

It

0.57

It

0.55

Ironically

0.47

Interestingly

0.46

不思議

0.45

 effectivement

0.45

 inderdaad

0.45

Actually

0.44

 Ironically

0.43

Activations Density 0.006%

punctuation marks

introduces an assertion or observation

No Comments

No Known Activations